English
Related papers

Related papers: MultiPath Memory Access: Breaking Host-GPU Bandwid…

200 papers

Multi-accelerator servers are increasingly being deployed in shared multi-tenant environments (such as in cloud data centers) in order to meet the demands of large-scale compute-intensive workloads. In addition, these accelerators are…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-10-08 Kiran Ranganath , Joshua D. Suetterlein , Joseph B. Manzano , Shuaiwen Leon Song , Daniel Wong

Training large language models (LLMs) in the cloud faces growing memory bottlenecks due to the limited capacity and high cost of GPUs. While GPU memory offloading to CPU and NVMe has made large-scale training more feasible, existing…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-19 Sabiha Afroz , Redwan Ibne Seraj Khan , Hadeel Albahar , Jingoo Han , Ali R. Butt

In this paper, we propose bit-interleaved multiple access (BIMA) to enable Internet-of-Things (IoT) networks where a massive connection is required with limited resource blocks. First, by providing a true power allocation (PA) constraint…

Information Theory · Computer Science 2023-04-13 Ferdi Kara , Hakan Kaya , Halim Yanikomeroglu , Chan-Tong Lam , Ben K. Ng

The two main thrusts of computational science are more accurate predictions and faster calculations; to this end, the zeitgeist in molecular dynamics (MD) simulations is pursuing machine learned and data driven interatomic models, e.g.…

Computational Physics · Physics 2020-02-24 Saaketh Desai , Samuel Temple Reeve , James F. Belak

The proliferation of camera-enabled devices and large video repositories has led to a diverse set of video analytics applications. These applications rely on video pipelines, represented as DAGs of operations, to transform videos, process…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-05-31 Francisco Romero , Mark Zhao , Neeraja J. Yadwadkar , Christos Kozyrakis

Memory tiering provides a cost-effective solution to increase memory capacity, utilization, and even bandwidth. Memory tiering relies on system software for memory profiling, detection of frequently accessed pages, and page migration. Such…

Operating Systems · Computer Science 2026-04-15 Xi Wang , Jie Liu , Shuangyan Yang , Jongryool Kim , Pengfei Su , Dong Li

Memory and logic integration on the same chip is becoming increasingly cost effective, creating the opportunity to offload data-intensive functionality to processing units placed inside memory chips. The introduction of memory-side…

Hardware Architecture · Computer Science 2017-08-23 Javier Picorel , Djordje Jevdjic , Babak Falsafi

It is often said that one of the biggest limitations on computer performance is memory bandwidth (i.e."the memory wall problem"). In this position paper, I argue that if historical trends in computing evolution (where growth in available…

Operating Systems · Computer Science 2011-05-11 Niall Douglas

Distributed ML workloads rely heavily on collective communication across multi-GPU, multi-node systems. Emerging scale-up fabrics, such as NVLink and UALink, enable direct memory access across nodes but introduce a critical destination-side…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-06 Amel Fatima , Tuan Ta , Bradford M. Beckmann

CUDA Unified Memory improves the GPU programmability and also enables GPU memory oversubscription. Recently, two advanced memory features, memory advises and asynchronous prefetch, have been introduced. In this work, we evaluate the new…

Distributed, Parallel, and Cluster Computing · Computer Science 2020-03-03 Steven W. D. Chien , Ivy B. Peng , Stefano Markidis

PIM architectures aim to reduce data transfer costs between processors and memory by integrating processing units within memory layers. Prior PIM architectures have shown potential to improve energy efficiency and performance. However, such…

Hardware Architecture · Computer Science 2025-10-10 Parker Hao Tian , Zahra Yousefijamarani , Alaa Alameldeen

This work introduces, for the first time, non-orthogonal multiple access (NOMA) into short-packet communications to achieve low latency in wireless networks. Specifically, we address the optimization of transmission rates and power…

Information Theory · Computer Science 2017-09-20 Xiaofang Sun , Shihao Yan , Nan Yang , Zhiguo Ding , Chao Shen , Zhangdui Zhong

Future 5G communication systems require more demanding performances than the existing cellular communication systems, e.g., 10 to 100 Mbps user data rate and much larger cellular spectrum efficiency. The well-used multiple access methods…

Signal Processing · Electrical Eng. & Systems 2020-02-04 Shin-Yuan Wang , Wei-Han Hsiao , Kang-Lun Chiu , Chia-Chi Huang

Various verticals in 5G and beyond (B5G) networks require very stringent latency guarantees, while at the same time envisioning massive connectivity. As a result, choosing the optimal multiple access (MA) technique to achieve low latency is…

Signal Processing · Electrical Eng. & Systems 2020-01-30 Mouktar Bello , Wenjuan Yu , Mylene Pischella , Arsenia Chorti , Inbar Fijalkow , Leila Musavian

End-to-end Vision-Language-Action (VLA) models for autonomous driving unify perception, reasoning, and control in a single neural network, achieving strong driving performance but requiring 20-60GB of GPU memory-far exceeding the 12-16GB…

Artificial Intelligence · Computer Science 2026-05-13 Seungwoo Roh , Huiyeong Kim , Jong-Chan Kim

The Lattice Boltzmann method (LBM) for solving fluid flow is naturally well suited to an efficient implementation for massively parallel computing, due to the prevalence of local operations in the algorithm. This paper presents and analyses…

Distributed, Parallel, and Cluster Computing · Computer Science 2017-01-04 Mark Mawson , Alistair Revell

Cutting-edge embedded system applications, such as self-driving cars and unmanned drone software, are reliant on integrated CPU/GPU platforms for their DNNs-driven workload, such as perception and other highly parallel components. In this…

Distributed, Parallel, and Cluster Computing · Computer Science 2020-03-20 Soroush Bateni , Zhendong Wang , Yuankun Zhu , Yang Hu , Cong Liu

The torrential influx of floating-point data from domains like IoT and HPC necessitates high-performance lossless compression to mitigate storage costs while preserving absolute data fidelity. Leveraging GPU parallelism for this task…

Databases · Computer Science 2025-11-12 Zheng Li , Weiyan Wang , Ruiyuan Li , Chao Chen , Xianlei Long , Linjiang Zheng , Quanqing Xu , Chuanhui Yang

Many performance critical systems today must rely on performance enhancements, such as multi-port memories, to keep up with the increasing demand of memory-access capacity. However, the large area footprints and complexity of existing…

Hardware Architecture · Computer Science 2020-01-28 Hardik Jain , Matthew Edwards , Ethan Elenberg , Ankit Singh Rawat , Sriram Vishwanath

We present the first comprehensive study of latent multi-head attention (MLA) for small language models, revealing interesting efficiency-quality trade-offs. Training 30M-parameter GPT models on 100,000 synthetic stories, we benchmark three…

Computation and Language · Computer Science 2025-06-17 Sushant Mehta , Raj Dandekar , Rajat Dandekar , Sreedath Panat